is the difference between the literature values and the experimental correlations greater than 1%. The mixture vapor pressure data of this study are approximately 1% greater than those reported by Manley and Swift a t the 100°F isotherm. This difference is attributed t o the difference in the method of pressure measurement employed with the modified equipment. The computed relative volatilities are compared with the literature data in Figures 9-13. The data of the previous investigators were acquired by sampling liquid and vapor at equilibrium in all but Figure 13. The data of Mann et al. in Figure 13 were taken in a n equilibrium still and exhibit less scatter than other data previously reported and show excellent agreement with the computed values of this study.
SUPERSCRIPTS L = liquid phase V = vapor phase SUBSCRIPTS 1 = propane 2 = propene i = component i r = pressure subscript, P computed using Raoult's law LITERATURE CITED
Beattie, J. A., Poffenberger, N., Cenfield, H., J. Chem. Phys., 3, 96 (1935).
Chernev. B. J.. Marchman, H.. York, Robert, Ind. E m . I
The relative volatilities reported in this akticle extend the range of information on the propane-propene system to the temperatures of greatest industrial interest. Since most industrial separations of propane and propene are carried out at temperatures in excess of 100"F, the data of this study should provide a useful extension of the data reported by Manley and Swift ( 8 ) . The combined data provide relative volatilities for the propane-propene system from -20" to f160"F. NOMENCLATURE
ABCD P T T, V
vapor pressure correlation constants pressure, psia temperature, "F temperature, OR specific volume, ft3/lb-mol 2 = compressibility factor, P V / R T , z = mole fraction propene in liquid y = mole fraction propene in vapor = = = = =
GREEKLETTERS a = relativevolatility [y(l - z)]/[z(1 - y ) ] A = difference
,
Chem., 41, 2653 (1941). Deschner. W. W.. Brown. G. G.. ihid.. 32. 836 (1940). Farrington, P. S.,'Sage, B: G., ihid., 41, 1734 (1949). ' Hanson, G. H., Hogan, R. J., Melson, W. T., Cings, M. R., ihid., 44,604 (1952). Laurance. D. R.. MS thesis. University of Kansas, Lawrence, KS, 1971: Manlev. D. B.. PhD thesis. ihid.. 1970 Manle;: D . B.'. Swift. G. W.. J.'Chem. Enu. Data., 16,. 301-7. ( 1971): Mann, A. N., Pardee, W. A., Smyth, R. W., ibid., 8 , 499502 (1963). . , Marchman, H., Prengle, H. W., Motard, R. L., Ind. Eng. Chem.,41,2658 (1949). Powell, T. M., Giauque, W. F., J. Amer. Chem. Soc., 61, 2366 (1939). Reamer, H. H., Sage, B. H., Ind. Eng. Chem., 43, 1628 (1951). Reamer, H. H., Sage, B. H., Lacey, W. N., ibid., 41, 482 ( 1949'). Sage, B. H., Schaafsma, J. G., Lacey, W.N., ibid., 26, 1218 (1934). Seibert, F. X., Currell, G. A,, J. Amer. Chem. SOC.,37, 2683 (1916). Vaughn, W. E., Graves, N. R., Ind. Eng. Chem., 32, 1252 (1940).
CONCLUSIONS
RECEIVED for review November 13, 1971. Accepted February 28, 1972.
Temperature Dependence in Determination of Solubilities in System Methylvinylketone in Water JAN VOJTKO' and MlLlNA CIHOVA Faculty of Chemical Technology, Slovak Technical University, Bratislava, Czechoslovakia
The dependence of solubility in the system methylvinylketone-water on temperature was measured using two mutually independent methods. These two substances form a partially miscible system. The immiscibility range i s limited by the lower (27.9"C) and upper (83.0"C) critical temperatures of solubility and by the concentrations 34.7 and 73.7 wt % of methylvinylketone in water.
Methylvinylketone (MVK), a technologically important intermediate, may be prepared by various methods. The most usual methods of preparation are based on the hydrolysis of cisor trans-1,3-dichloro-2-butenewith N a ~ C 0 3and HzS04 (4), or catalytically with CuClz (3) or with HCl ( 8 ) . It is also pos1
To whom correspondence should be addressed.
sible to prepare it by condensation from acetone and formaldehyde ( I , 5, 6, IO), by dehydrogenation of methylethylketone (?'), or from vinyl magnesium chloride and acetylchloride (9). One of the oldest methods is its preparation by hydration of vinylacetylene ( 2 ) . It is necessary to know the mutual solubility of methylvinylketone and water for this procedure. This paper presents data as to this solubility.
Journal of Chemical and Engineering Data,
Vol. 17, No. 3, 1972
337
EXPERIMENTAL
Methylvinylketone (National Enterprise Duslo, Sal'a), prepared by the hydration of vinylacetylene, had the following constants: Reagents.
bp?,, = 81.3OC n20D =
bp,,
=
Table II. Temperature Dependence of Solubilities in System Metylvinylketone-Water as Measured Refractometrically
81.4OC (IO)
n% = 1.4110 (4)
1.4114
dd20 =: 0.8494
dh2' = 0.8501
Equilibrium temp
(4)
27.9 28.5 30.7 33.3 36.7 42.0 48.0 52.0 62.0 66.0 70.0 74.6 80.0 83.0
The water was redistilled.
APPARATUS AND PROCEDURE
The solubility determination in the system methylvinylketone-water is difficult because methylvinylketone polymerizes very easily just as do other a,p-unsaturated ketones. Since polymers, even in minute amounts, have a substantial effect on the mutual solubility of the two components of the system, we selected two independent methods for the solubility measurement to increase the accuracy of determination and objectivity of the measurement. The system was investigated by both cloud point measurement and by refractometry. In the first case, we observed the onset of turbidity in sealed ampuls containing mixtures with known concentrations; in the second case, the phases were equilibrated a t a given temperature and both phases then analyzed refractometrically. The composition was then read from a calibration plot prepared by determining the refractive indices of known solutions. In the cloud point determinations, an optical glass vessel was used as the thermostat. It was equipped with a stirrer, temperature was controlled to 0.05"C by Anschutz thermometers (0.l"C calibrations), and a stand was provided for holding the sealed ampuls. For the refractometric method, we used test tubes having a 1 5 m l volume and equipped with a stopcock. The samples were taken using a syringe with a long needle. The analysis was made refractometrically on an Abbe refractometer with thermally controlled prisms.
Compn of organic layer, wt % MVK
Compn of
aqueous layer, w t 7% MVK
in water in water Appearance of the first turbidity 63.2 67.0 69.4 71.4 72.9 73.5 73.2 71.5 70.6 68.5 67.3 60.1
50.0 46.9 42.6 38.8 36.5 36.1 35.7 35.0 34.8 35.2 35.6 37.7
Existence of one phase
obtained using the refractometric method are listed in Table 11. The phase diagram of solubility of the system and its dependence on temperature as measured by both methods is shown in Figure 1. The lower and upper critical solubility temperatures, 27.9' and 83.0°C, as well as the concentration ranges limiting the region of heterogeneity, were determined graphically from Figure 1. The limits of this range are a t 34.7 and 73.7 w t %, respectively, of methylvinylketone in water. DISCUSSION
Figure 1 shows that the solubility dependence of the system methylvinylketone-water on temperature has an unusual shape. The solubility decreases a t first with increasing temperatures, but then it increases again. The phase diagram of the mea-
RESULTS
The results of the solubility determination obtained by the cloud point method are summarized in Table I. The results
Table I.
Temperature Dependence of Solubilities in System Methylvinylketone-Water as Determined by Cloud Point Measurements
Concn, wt % MVK 32.4 35.0 38.9 40.1 42.6 45.5 47.9 50.5 52.4 55.0 57.1 61.3 64.2 67.5 70.2 72.8 74.9
338
'C
t 70 .
Cloud point temp Lower, "C Upper, "C
...
...
50.5 36.8 35.5 33.8 30.5 29.6 28.4 28.2 28.0 28.0 28.1 28.5 30.3 33.5 41.5
69.5 82.5 82.9 82.8 82.5 82.2 82.0 81.7 81.1 79.5 77.3 74.0 70.5 65.1 54.0
..
...
Figure 1 .
Journal of Chemical and Engineering Data, Vol. 17, No. 3, 1 9 7 2
0
Solubility of methylvinylketone in water
= Refractometric data.
V =
Cloud point data
sured system forms a closed curve between the upper and lower critical limits of solubility limited by the temperatures 27.9" and 830°C and by the concentrations 34.7 and 73.7 wt %, respectively. The determination of solubility using two methods (cloud point and refractometry) has given identical results except for the methylvinylketone phase a t higher temperatures (refer to Figure 1, hatched section) where the refractometer (circles) indicated a higher methylvinylketone content and therefore a lower water solubility in the methylvinylketone phase. Although measurements by the cloud point method are commonly more accurate (it is possible to weigh the samples into the sealed test tubes very precisely by differential weighing, the lower limit of turbidity detection is less than 0.1"), it is less suitable than refractometry for this particular system because of the polymerization activity of the methylvinylketone. Even with 0.05 w t % of diethylhydroxylamine added as stabilizer, a partial polymerization occurs a t higher temperatures, as the sample is exposed to the influence of heat and light a rather long time (on the order of several hours). Since polymers affect the cloud point considerably, even in small amounts, the authors tend to the opinion that the refractometric method of determination gives more truthful results. The same concentration of polymers affects the
change of cloud point more than the refractive index of the mixture. Besides, the possibility of an error in the refractometric method was decreased by changing the samples for each temperat,ure measurement, and the calibration measurements were made with mixtures containing only pure methylvinylketone without polymers. LITERATURE CITED (1)
Brant, J. H., Hasche, R. L., U. S. Patent 2,245,567 (June 23, 1939).
(2) (3)
Carter, A. S., ibid., 1,896,161 (Nov. 11, 1930). Carter, A S., Johnson, F. W., ibid., 2,263,379 (Aug. 11,
(4) (5)
Churbakov, A N., Zh. Obshch. Khim.,10,977 (1940). Churbakov, A. N., Ryazantsev, V. N., Zh. Prikl. Khim.,
1939).
13, 1964 (1940).
Decombe, J., Compt. Rend., 202, 1685 (1936). Kalb, G. H., U. S. Patent 2,719,171 (Sept. 27, 1955). (8) Kalina, K., Czechoslovakia Patent 102,024 (July 12, 1960). (9) Ramsden, H. E., U. S. Patent 2,873,275 (Feb. 10, 1959). (10) White, T., Haward, R. N., J. Chem. SOC.,1943, p 25. (6) (7)
RECEIVED for review November 17, 1971. Accepted March 20, 1972.
Viscosity Studies of Certain Alkyl Phosphates and Their Aqueous Mixtures GABRIELLA THAU-ALEXANDROWICZ and GABRIEL MICHAELI' Polymer Department, The Weizmann Institute of Science, Rehovot, Israel
The temperature- and water-concentration-dependent kinematic viscosity of the phosphate esters, tri-n-butylphosphate (TBP), di-n-butylcresylphosphate (DBCP), dicresyl-n-butylphosphate (DCBP), tetra-butylhydroquinonediphosphate (HQDP), tetra-butylbutylenediphosphate (TBBDP), and their water mixtures, was measured by use of water-calibrated Ubbelhode-Cannon viscometers within the temperature range 0 4 5 ° C . The results are well represented by the relationship v ( T ) exp ( A 6/T C / T Z ) in which the activation energy has the form EJT) A R(6 C / T ) . These results are discussed with relation to the solvents' self-association and H-bonding properties.
= +
Solvent membranes made of tri-n-butylphosphate (TBP) exhibit highly selective water transport properties ( I S ), attributed to the formation of TBP-water complexes in the organic phase ( 7 ) . Similar effects were observed in several other organophosphate solvent membranes examined in our laboratory ( l J6J1 2 ) *
The viscosity study reported here is part of a project aimed a t determining the molecular mechanisms involved in the water transport properties of these solvent membranes. EXPERIMENTAL SECTION
Materials. Pure tri-n-butylphosphate (TBP) was obtained from Fluka A.G., di-n-butylcresylphosphate (DBCP) (bp 132-3"C/0.3 mm Hg) , dicresyl-n-butylphosphate (DCBP) (bp 18OoC/2.5 mm Hg) tetrabutylhydroquinone diphosphate ~
1 Present address, Department of Atmospheric Sciences, The Hebrew University of Jerusalem, Israel. To whom correspondence should be addressed.
=
+
+ +
(HQDP) (bp > 100°C/10-3 mm Hg) and tetrabutylbutylenediphosphate (TBBDP) (bp > 100°C/10-3 mm Hg) were synthesized a t the Plastics Research Laboratory, Weizmann Institute by Sylveta Marian and David Vofsi. The purity of the phosphate esters was estimated a t >97% by comparing their empirical phosphorus content to the calculated one. All solvents were dried in a vacuum desiccator heated to 60°C and containing solid NaOH for a t least 24 hr before use. Methods. Densities were measured with an A. P. Gratz digital densitometer. Water concentrations were determined by the Karl Fischer method. The kinematic viscosity of the organophosphates and their water mixtures was measured with water-calibrated Cbbelhode-Cannon viscometers, within the temperature range 0-45°C. The temperatures were kept constant within +=O.O5"C during each measurement. The precision of the viscosity measurements was estimated a t < =k0.2'%. For checking the viscosity water-concentration dependence, a measured volume of distilled water from an Agla microsyringe was added after each viscosity run, into the weighted solvent sample being measured. Complete mixing of the organophosphates' water mixtures was
Journal of Chemical and Engineering Data, Vol. 17, No. 3, 1972
339
